skill-creator

To package modular AI skills, you structure them using specified frontmatter, organize resources into scripts, references, and assets, and apply progressive disclosure to ensure consistent, safe, and reusable units.

Progressive disclosure in AI skill design is a structuring approach that reveals information and capabilities in layers, ensuring the skill remains consistent and safe while staying modular and reusable across domains.

You scaffold a new skill by running the skill initializer to generate the base structure, then customize the SKILL.md file and bundled resources to update existing ones or extend Assistant capabilities.

Yes, pyyaml is a required dependency for the skill-creator framework, enabling the processing and validation of frontmatter structures specified within your SKILL.md files.

The best way to design frontmatter for AI skill templates is to follow the specified frontmatter structure provided in the skill-creator references, ensuring metadata is properly defined for consistent packaging and execution.

When updating existing skills across domains, limitations may arise if resource organization deviates from the standard scripts, references, and assets structure, potentially breaking the progressive disclosure needed for safe reuse.